Type II errors are false negatives. They occur when a predictive model or statistical test fails to identify an effect that actually exists. For example:
- In a survey comparing the performance of Business Development Representatives (BDRs) with and without a sales engagement platform (SEP), a Type II error occurs if the statistical test fails to detect a meaningful difference in performance between BDRs with and without SEPs, even though such a difference exists.
Often, small sample sizes will cause a statistical test to return a non-significant result, even if there is a meaningful difference. As a result, small sample sizes are often the cause of false negative (Type II) errors.
In typical business situations, it is up to the practitioner or researcher to determine how much risk is acceptable. This means weighing the consequences of a Type I versus a Type II error. It is always a balance of the two.
